Digital Subscriber Line (DSL) is a broadband access technique. DSL has gained wide popularity in hundreds of thousands of households. A typical DSL networking is illustrated in FIG. 1. The extensive applications of DSL technique, such as Voice over IP (VoIP), IP Television (IPTV), have broadened the application field for broadband technology. Various IP based devices have come into existence based on these applications or combination thereof. The IP application terminals now come into use in the households are called Customer Premises Equipments (CPEs), which are configured automatically by an Auto-Configuration Server (ACS). Accordingly, the CPE may operate normally in absence of a user's configuration.
A series of technical documents proposed regarding CPE specify a data model supported by CPE and parameters that the CPE must support and parameters that the CPE may support by option. For a particular type of CPE, such as a particular model from a particular manufacturer, a particular software version and one or more type of the particular hardware version, the data model of those may be the same. To operate a specific CPE, the ACS has to know about the data model of the CPE and a Remote Procedure Call (RPC) method supported by CPE so that it may operate and configure the CPE without error.
However, there are no good solutions so far to handle the issue as to how an ACS obtains the data model supported by the CPE and the RPC method supported by CPE as well as how the ACS associates a specific CPE with the specific Data model supported by the specific CPE and RPC method supported by CPE on the ACS. Currently, the ACS employs a GetParameterNames RPC method to directly obtain parameters supported by the CPE and the read/write property of these parameters from the CPE. However, such an RPC method is confronted with two issues.
The first issue is that the data model obtained using such method is incomplete. The data model does not include the type and range of the parameters supported by the CPE.
The second issue is that, according to the method, the ACS has to obtain the data model of each CPE every time the ACS operates a CPE. Even though these CPEs have the same data model, the workload of the ACS may be increased.